PLANKTON 239 



since the days of P. T. Cleve and from the observations of 

 Hensen's own colleagues that this is not the case, and they 

 have pubhshed chart-diagrams showing that at least three 

 different kinds of water under different conditions are found 

 in the North Sea and that at least five different planktonic 

 areas may be encountered in making a traverse from Germany 

 to the British Isles. 



There is also direct evidence of irregularity in the dis- 

 tribution of such fish eggs. Hjort and Petersen, in 1905, 

 showed that cod eggs are found in great quantities over the 

 isolated banks of the coast of Norway, while none or very 

 few are found over the channels between the banks. Schmidt 

 also found eggs and fry of cod on the RockaU Bank, but not 

 outside it. If the argument be used that wherever the 

 plankton is found to vary, there the conditions cannot be 

 uniform, then few areas of the ocean of any considerable size 

 remain as cases suitable for population-computation from 

 random samples. 



The Kiel School of Planktologists cannot have it both ways. 

 They claim that the adequacy of their samples holds good 

 for an area of sea all of which is under similar conditions. 

 They teU us at one time that the North Sea contains water of 

 different kinds from dift'erent sources and with several types 

 of plankton. If, then, it is not homogeneous — as of course, 

 from aU the evidence, it is not — then they cannot average 

 the samples and multiply up for the whole area as Hensen 

 and Apstein have done. 



We have pubhshed many examples from the Irish Sea of 

 marked irregularity in the plankton. If the plankton were 

 uniformly distributed, then two ordinary open horizontal 

 nets towed together at the same time ought to show similar 

 catches, and they sometimes do ; but very often they do 

 not. Even when the volume of the catch is much the 

 same in a pair of nets, the totals may be made up very 

 differently, as in the case of nets A and B shown in the 

 table on next page. 



